Toxicities Associated with Anti-PD-1 Immunotherapy

Transcript:Mark Socinski, MD: The tolerability speaks to the rate of toxicities that we see with this. And David mentioned the difference between the rates of toxicities with chemotherapy versus pembrolizumab in KEYNOTE-024. Jared, first, the spectrum of toxicity is very different than chemotherapy. There’s a trend here, right? When the EGFR TKIs came out, we had to be dermatologists. When anti-VEGF agents came out, we had to be the cardiologist and then the internist to manage hypertension. Now we’ve got to become endocrinologists with all these -itises, in terms of what they incurred.

Tracey L. Evans, MD: And the rheumatologists.

Mark Socinski, MD: Yes, rheumatologists. Give us an overview of the spectrum of toxicities we see with immune agents, and then we’ll talk in a little bit more detail about some of these.

Jared Weiss, MD: As you pointed out, while these agents may be less toxic than doublet chemotherapy, they are not toxicity-free and they are not risk-free. The most common autoimmune effect that we’re seeing from this class of agents is endocrine. And within those, as was alluded to, hyperthyroidism and hypothyroidism are high on the list of what we’re seeing most often. In my practice, I do monitor patients for thyroid function. I give a TSH every few treatments. And these are usually easily manageable with levothyroxine or carbimazole, depending on which direction you’re going in.

For other endocrine disruptions, you sometimes might need consultation with an endocrinologist. And certainly for those in the early goings of managing these, I would encourage a very low threshold to use such consultation. You can see hyperglycemia that is and is not type 1 diabetes. So, you can see a frank autoimmune type 1 diabetes-type picture or you can just see high sugars. And pretty much anywhere else in the endocrine access, someone has reported a mild or severe toxicity, the most feared being hypophysitis.

On the more severe end of the spectrum, patients, particularly those with a heavy smoking history, are at risk for pneumonitis. We’re talking about 1% or 2% of patients, but this could be pretty severe. Patients can be very short of breath, they can be hypoxic. There’s a classic imaging appearance that can help when there’s a doubt with this. These need to be recognized very promptly. These are ones you need to warn patients about that if they get more short of breath. They need to come in quickly, you need to get imaging rapidly. And the gist of this is that you need to get them on high-dose steroids rather quickly. Perhaps 1 mg/kg or so of prednisone, and they need that to be tapered very, very slowly. In some cases, a pulmonary consultation can be very helpful.

We also fear autoimmune colitis, although this seems to be rarer than with the CTLA4 experience. Again, the first-line treatment is high-dose steroids and stopping the agent while they’re on it. And then, there are rarer toxicities that are seen. I think that when those are seen, it’s helpful to speak with someone who has seen a lot of toxicities from these drugs, who has used them a lot either in practice or in trials, and to bring in the relevant subspecialist to assist with their management.

Mark Socinski, MD: The thing that I think other physicians need to be aware of, too, is that when you look at the time-to-onset of these toxicities, most of them occur relatively early.

Jared Weiss, MD: But not all?

Mark Socinski, MD: Yes. These are things that can occur at any point. And I used to emphasize to the Fellows in Pittsburgh that oftentimes these patients tolerate these agents extremely well. It’s almost like you let your guard down because Mrs. Jones is doing so well. Then, Mrs. Jones starts to have something that’s grade 1 that gets a little blown off, and then all of a sudden Mrs. Jones has a grade 3 toxicity that could have been recognized earlier and dealt with earlier with appropriate supportive care, including steroids.

Jared Weiss, MD: Absolutely. And that early supportive care is not just because you don’t want Mrs. Jones to suffer at the moment, although that’s a good enough reason to bring her in and give her high-dose steroids. There’s good literature from melanoma that we’re starting to explore more in lung cancer to show that if you do aggressively treat that early, some of these patients may be able to get back on the drug and benefit.

Mark Socinski, MD: And in some patients that I’ve seen in consultation that didn’t go back to their primary oncologist, I’ve had a few oncologists say to me that they withheld steroids for the fear of blunting any sort of immune response. Tracey, have you heard this?

Tracey L. Evans, MD: Well, yes. That’s a fear of mine, too. A lot of the studies excluded people on steroids. But if they’re having one of these autoimmune events, you absolutely have to treat it. You need to get it under control or it can be very dangerous. There have been fatal episodes of pneumonitis.

So, what’s amazing to me with these drugs is there’s no predictable toxicity. I know when I give cytotoxic chemotherapy, they’re going to have some nausea and vomiting, especially around the antiemetics. They’re going to have some fatigue; their counts are going to go down. But many people getting these agents have absolutely no toxicity. You don’t need to give any supportive care drugs, which is just so amazing when you’re writing the chemotherapy orders and it’s just the drug. And yet, you’re going to get these random rare toxicities that you’ve never seen before. I had one patient who had this very severe headache and an aseptic meningitis picture, which, yes, it can happen. I had another patient who had a polymyalgia rheumatica-type syndrome, which makes sense; that’s immune-based. His joints really ached and he needed to stay on a low-dose prednisone, yet it seemed like that didn’t impact the drug’s ability to work.

Mark Socinski, MD: In all of these trials, patients with prior autoimmune diseases were excluded. I think I saw at ASCO last year that if you look in the SEER database, about 25% of the lung cancer population would have something that one could consider a history of autoimmune diseases, and they probably were excluded from this. Many of these are not well-documented and may be very remote. What do you do in these situations? Do you ignore them? Or what advice do you have for community doctors?

Tracey L. Evans, MD: For me, I eventually ignore them, meaning I’m not going to let somebody with lung cancer die of their disease because of some autoimmune disease without having seen a checkpoint inhibitor. Where I might use it will depend on how severe whatever disease they have is, how concerned I am about it.

Jared Weiss, MD: The line of therapy.

Tracey L. Evans, MD: Exactly. So, I might not use pembrolizumab first-line in spite of high expression if somebody has just really severe lupus; not like with rheumatoid arthritis, but if there’s some severe conditions. If for their colitis that they already have, I might be more reluctant to use it. But they’ll get it eventually, and I’ll use steroids if I need to.

Jared Weiss, MD: Our melanoma colleagues who’ve been using these drugs—at least checkpoint inhibitors, in general, for a few years longer than us—have, over time, reported that they’re relaxing those thresholds. And they largely get away with it. One place that I’m less tempted to try to get away with it is ILD (interstitial lung disease), where people seem to get into trouble at a much higher rate.

Mark Socinski, MD: So, having said that, the next question I was going to ask is, would you rechallenge after an immune-related adverse event? You’re hinting that with an ILD-type, you probably wouldn’t rechallenge.

Jared Weiss, MD: I would not.

Tracey L. Evans, MD: Preexisting ILD or someone who gets pneumonitis on the drugs?

Mark Socinski, MD: No, no, someone who had pneumonitis.

Jared Weiss, MD: So, that’s a really important distinction. Someone with preexisting ILD, I’m certainly pushing this to later lines, if ever at all. That’s a high note of caution for me. I have already had, even outside of trials, a standard-of-care use for patients who have had pneumonitis, including at least one patient with rather dramatic pneumonitis. They got high-dose steroids, got better from it, have gotten back on drug and have benefited. And that has been reported more extensively in the literature. That’s very real.

Tracey L. Evans, MD: And I’ve done that, too. But what gives me some pause is the fact that in patients who’ve had to go off these agents because of toxicities, they can have a durable benefit, even without going back on it. So, I do feel nervous about taking that risk.

Mark Socinski, MD: Which brings up one of the issues that I think we have to deal with at some point—not during this, but at some point in clinical research—and that is duration of therapy with these agents because they are very costly. And if we can get away with shorter durations, then it would, with the same outcome, be helpful.

Ed, I want to get back to you. I want to ask you the question. I’ve heard theoretical arguments as to why PD-L1 inhibitors may be less toxic than PD-1 inhibitors. The severe toxicities occur so uncommonly that it’s hard to cross trials, say that one is clearly better than the other. Do you have any sense or do you just see them all as the same for the side effects?

Edward S. Kim, MD: I think we’re still very nascent in this, and we rely on one of our immunotherapy doctors who does melanoma and renal cell and who did high-dose IL-2 before. You want some side effects, so go to a high-dose IL-2 program. So, we rarely have those patients admitted any more to the ICU beds like we used do. But when you use their expertise, because they have been dealing with this, I think when we do this next year or when we have topics next year, we’re going to be talking about the combinations. And this isn’t nothing yet. Just wait for those. We’re dabbling in the minor leagues right now, according to them, because you start adding on the additional agents on top of these, whether it’s CTLA4 or PD-1, PD-L1. The IOs seem to be less as far as some of them, but I think it’s going to definitely take a different shape and a different form, even as we start looking at combinations with chemotherapy.

I do think we should hopefully learn the lessons that we have with other eligibility criteria. And I think an important one that you brought up is brain metastases. A 15-year-old versus an 18-year-old, this is a movement we’re trying to push out there. That a 12-year-old metabolizes drugs the same way as an 18-year-old. Yet for some reason, it’s something magic when you’re 18 and you get to be on a clinical trial or treatment and others don’t.

So, hopefully we will be more receptive to change, especially with the history of some of these autoimmune diseases, to treat patients and not leave people out in the dark. Because right now, those community doctors, they’re hearing some different messages here and there. Some of them are not using steroids. I absolutely believe in stopping, giving the steroids that you need, getting them tapered down, and getting them restarted. I personally had more issues with elevated liver function tests, which is very frustrating because the patient looks fine in front of you, they’re not short of breath, they’re not having weird thyroid symptoms or anything, and their LFTs are just staying up.

Mark Socinski, MD: Yes. You’re treating a number.

Edward S. Kim, MD: That’s right. And it’s very challenging, because I’ve gone through this with one patient where we’ve tried actually 2 different immunotherapies now, and both times there was a bump in the LFTs. And this is where I get scared. I totally believe that there is a lasting effect when you have a response. This, of course, happened to her. She had a response and then her LFTs bumped up. I said, “Don’t worry, it’s around, it’s around.” We restaged her, and it’s grown and doubled in size, of course. We had to get her back on a different one right away. So, it’s going to always be the one.

Tracey L. Evans, MD: She re-responded?

Edward S. Kim, MD: We’re in the process of looking at it right now.

Transcript Edited for Clarity